INDEX

Explanations

references to legal or criminal consequences

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

(Void

-0.15

arella

-0.14

à¸Ħà¸£à¸ļ

-0.14

Î»Ïī

-0.14

.addProperty

-0.14

çī¹èī²

-0.14

Ø§Ø¨ÙĦ

-0.13

_Ptr

-0.13

ì¼ĢìĿ´

-0.13

Gratis

-0.13

POSITIVE LOGITS

 plot

0.39

 plots

0.34

plot

0.34

 plotting

0.33

 conspir

0.32

 Plot

0.31

Plot

0.29

 conspiracy

0.29

 cons

0.29

 comp

0.29

Activations Density 0.291%

references to legal or criminal consequences

No Comments

No Known Activations

references to legal or criminal consequences

No Comments

No Known Activations